There are a number of people who have consistently initiated projects that failed to deliver the promised outcome, and cost a lot of money and other negative impacts. Yet through clever promotion these people continue to be in a position to repeat their failures.
People also gain skill over time. They learn from mistakes, and failure is not always a predictor of future failure. Also, the mantra of “fail early, fail often” has proven to be effective in many domains. Consequences of failure need to be minimised, while balanced with avoiding future failures.
Also people who have great potential in predicting the success of a project may not always be in a position to influence whether a project goes ahead or not. This failure to use available skills and knowledge results in avoidable project failures.
A solution that objectively measures how likely a person is to be correct when predicting the outcome of a project could substantially impact the number of successful and failed projects in a very positive way.
All projects should include what benefits they will provide, how much they will cost, and how long they will take to deliver. Without this information, a project cannot be judged as viable or not. These predictions should be related to Benefit points and based on the selected results of Outcome modelling
After completion, these outcomes should be documented stating the benefit they provided, their cost, and how long they took to complete.
Career points should enable people to bet on how close the predictions are to their outcomes. Participants should be able to bet a percentage of their career points, and be rewarded on how closely the project predictions aligned with the outcomes.
The total number of career points bet on a project should then be used to indicate the confidence in a project’s success. People directly involved in determining if a project should go ahead must bet a significant amount of their career points if they are recommending the project proceed.
Maybe only the people who have an influence on whether the project goes ahead or not should be able to bet on it. Maybe there is value in extending that out to a limited criteria of other people. Maybe it could be extended to everyone in some weighted manner. This needs to be determined. The decision should be based on what solution most increases the alignment between career points and successful projects.
Maybe a certain number of points are awarded in government when someone is promoted to a new pay grade. Maybe initiating new projects can be partially rewarded by providing some career points. Maybe each year some career points can be awarded based on pay grade, but not a significant enough number to disrupt the integrity of the system. Maybe there are other solutions too.
Career points should be bet by individuals as both a percentage of their total available points at the time of the bet, and the possibly a multiplier for how confident they are.
The project should already have clear predictions on benefit, cost, and duration, each expressed as a bell curve. The x axis represents the metric, such as cost. The y axis represents the probability of it being that metric and ranges from -1 to +1. This enable proponents to be clear about their confidence in the scope of these metrics.
On completion, the difference between the prediction and the outcome mark a point in that bell curve. Then career points are added or deducted based on:
Points bet x Confidence x y value on the bell curve
Assessing the total career points bet might also require a calculation of total points available to all participants and total confidence of all participants. This way a lot of participants betting a small amount of their points with low confidence will look like there is a low chance of project success, where as a small number of participants with high confidence and still indicate a successful project.